ProjectLawful.com: Eliezer's latest story, past 1M words

post by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-11T06:18:02.738Z · LW · GW · 112 comments

So if you read Harry Potter and the Methods of Rationality, and thought...

"You know, HPMOR is pretty good so far as it goes; but Harry is much too cautious and doesn't have nearly enough manic momentum, his rationality lectures aren't long enough, and all of his personal relationships are way way way too healthy."

...then have I got the story for you! Planecrash aka Project Lawful aka Mad Investor Chaos and the Woman of Asmodeus, is a story in roleplay-format that I as "Iarwain" am cowriting with Lintamande, now past 1,000,000 words.

It's the story of Keltham, from the world of dath ilan; a place of high scientific achievement but rather innocent in some ways.  For mysterious reasons they've screened off their own past, and very few now know what their prescientific history was like.

Keltham dies in a plane crash and ends up in the country of Cheliax, whose god is "Asmodeus", whose alignment is "Lawful Evil" and whose people usually go to the afterlife of "Hell".

And so, like most dath ilani would, in that position, Keltham sets out to bring the industrial and scientific revolutions to his new planet!  Starting with Cheliax!

(Keltham's new friends may not have been entirely frank with him about exactly what Asmodeus wants, what Evil really is, or what sort of place Hell is.)

This is not a story for kids, even less so than HPMOR. There is romance, there is sex, there are deliberately bad kink practices whose explicit purpose is to get people to actually hurt somebody else so that they'll end up damned to Hell, and also there's math.

The starting point is Book 1, Mad Investor Chaos and the Woman of Asmodeus. I suggest logging into ProjectLawful.com with Google, or creating an email login, in order to track where you are inside the story.

Please avoid story spoilers in the comments, especially ones without spoiler protection [LW · GW]; this is not meant as an "ask Eliezer things about MICWOA" thread.

112 comments

Comments sorted by top scores.

comment by Yoav Ravid · 2022-05-11T11:17:42.550Z · LW(p) · GW(p)

Is the forum format important, with the separation between posts and the profile pic and info on the side of each post? Cause if not I would love to have an epub version of this so I can read it on my kindle, might even find a way to create it myself if it doesn't exist yet.

Replies from: maia, KaynanK, rocurley, RomanS, Writer
comment by maia · 2022-05-11T16:43:19.338Z · LW(p) · GW(p)

https://github.com/rocurley/glowfic-dl

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-12T03:24:37.904Z · LW(p) · GW(p)

I liked how the epub strips out unnecessary UI from the glowfic site, but downloading and moving epubs around is a pain...

So I built a web reader on top of this code! Check it out here: https://share.streamlit.io/akrolsmir/glowflow/main

It'll work for any Glowfic post actually, eg https://share.streamlit.io/akrolsmir/glowflow/main?post=5111 Would probably be simple to add a download button to get the epub file; source code here.

Replies from: Eliezer_Yudkowsky, Raemon, Yoav Ravid, primerose
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-12T06:01:41.337Z · LW(p) · GW(p)

Oh, huh!  I can see that format potentially being preferred by a number of people.

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-12T12:24:45.031Z · LW(p) · GW(p)

I hope so! I myself bounced off of Mad Investor Chaos twice before because the formatting was too hard for me to read... but after implementing this reader, spent 2 hours last night reading through it.

Thanks so much for writing this Glowfic!

comment by Raemon · 2022-05-13T18:03:44.183Z · LW(p) · GW(p)

This is pretty great. One remaining issue is that it seems harder to hold onto your place if you stop in the middle of a section and then reload the page.

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T22:17:13.802Z · LW(p) · GW(p)

Thanks! I tried splitting into smaller sections (half the size) so that we don't have this issue as much; not sure what other solutions look like.

Replies from: Raemon
comment by Raemon · 2022-05-19T02:19:46.060Z · LW(p) · GW(p)

Ah thanks, that was helpful.

comment by Yoav Ravid · 2022-05-12T10:35:46.492Z · LW(p) · GW(p)

Nice!

I'd love a download option, but for now I'll use your reader to read it. I saw there's an option to make the text window wider, which is nice. Is there a way to make the text background dark?

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-12T12:23:27.556Z · LW(p) · GW(p)

Your wish is my command - dark mode added!

Replies from: Yoav Ravid
comment by Yoav Ravid · 2022-05-12T15:35:45.680Z · LW(p) · GW(p)

Awesome! Thanks!

One more thing, currently the box doesn't fully fit on my screen vertically when it's open (I'm using a laptop that doesn't have a high resolution). Is there a way to make the open box fit on the screen?

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-12T16:59:02.849Z · LW(p) · GW(p)

Hrm, I think I could code in a way to specify the height of the box... lemme look into it.

Thanks for all your suggestions, btw!

Replies from: gjm
comment by gjm · 2022-05-12T21:50:28.379Z · LW(p) · GW(p)

I have a kinda-opposite problem. There's a "wide mode" but not a "tall mode", and for reasons beyond my understanding the box with the actual content in it is about half the height of my browser window. I can understand why you wouldn't make it full-width by default -- very wide text can be hard to read -- but I can't think of any reason for preferring less than (full height minus reasonable margins).

Personally, for width I would prefer something intermediate between normal and "wide" (given the actual dimensions of my browser window, which occupies half of quite a large monitor) and for height I would prefer All The Pixels. Would it be difficult to have a "manual control" mode where you can set width, height and font size arbitrarily?

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T00:56:38.357Z · LW(p) · GW(p)

Done! Thanks for the feedback. Hoping 2000px is plenty but it's easy to increase lol.

(Having too many options is sometimes a symptom of bad UX design, but it seems reasonable for a web reader to support these all of these.)

Replies from: gjm, Yoav Ravid
comment by gjm · 2022-05-13T15:34:56.741Z · LW(p) · GW(p)

Nice!

I see it's some way behind the current state of the story. Is that because there's some background thing that fetches only every so often, or because there's a cacheing layer that might be stale, or what?

[EDITED to add:] I guess it's reading from the epub on Dropbox?

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T21:57:59.444Z · LW(p) · GW(p)

Yeah probably a stale caching layer, what fic were you reading? Glowflow doesn't read from an epub, it's reading html from the site itself.

Lemme try rebooting to see if that refreshes. That's obviously not sustainable... I didn't expect people to actually use it for a live, updating fic lol.

Edit: added a "Clear cache" button, hope that solves it!

Replies from: gjm, gjm
comment by gjm · 2022-05-14T18:52:01.171Z · LW(p) · GW(p)

Currently some way behind again; clicking "clear cache" does do something that looks like reloading but doesn't make it not be some way behind.

[EDITED a couple of days later to add:] No change to what the reader is showing since then even though there have been updates on glowfic.com. The "clear cache" button doesn't appear to help. Forcibly reloading the page also doesn't appear to help.

comment by gjm · 2022-05-13T22:26:29.695Z · LW(p) · GW(p)

I was, unsurprisingly given context :-), trying to read the latest Planecrash: https://share.streamlit.io/akrolsmir/glowflow/main?post=5930

Right now, what happens when I go there is that I get "Oh no. Error running app. If this keeps happening, please contact support." (In both Firefox and Chrome, on Linux. It was OK earlier today. I have Firefox configured to disable most Javascript but have whitelisted the domains I think the app actually needs, and again it was working OK earlier today, and Chrome doesn't have any of the JS-blocking stuff.)

The only possibly-useful-looking thing in the Javascript console says "INITIAL -> (10, 0, undefined) -> ERROR".

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T23:18:51.274Z · LW(p) · GW(p)

Sorry about that - had some configuration issues. It should work now!

Replies from: gjm
comment by gjm · 2022-05-13T23:34:50.777Z · LW(p) · GW(p)

Yup, all working (and apparently up to date) now. Thanks!

comment by Yoav Ravid · 2022-05-13T03:07:47.703Z · LW(p) · GW(p)

This is how it looks now when I increase the width. The outer box doesn't widen together with the text and background, and the text doesn't stay centered. Also, dark mode seems to have stopped working.

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T11:24:14.234Z · LW(p) · GW(p)

dark mode seems to have stopped working

 

Hm, do you have "dark mode" toggled on the sidebar? (There's two settings unfortunately due to how Streamlit is set up):

 

The outer box doesn't widen together with the text and background, and the text doesn't stay centered

Yeah unfortunately this is mostly working-as-implemented. The box size isn't a thing I can change; "Wide Mode" lets it the box be big, otherwise it's small.

Text centering might be possible if you're in "Wide Mode" -- I'll look into that.

Replies from: austin-chen
comment by Austin Chen (austin-chen) · 2022-05-13T12:26:16.738Z · LW(p) · GW(p)

Text centering should now be live!

comment by primerose · 2022-09-02T21:08:54.057Z · LW(p) · GW(p)

Late to the party but: I'd love an option to adjust line spacing!

comment by Multicore (KaynanK) · 2022-05-11T14:58:58.367Z · LW(p) · GW(p)

Somewhat. The profile pic changes based on the character's emotions, or their reaction to a situation. Sometimes there's a reply where the text is blank and the only content is the character's reaction as conveyed by the profile pic.

That said, it's a minor enough element that you wouldn't lose too much if it wasn't there.

On the other hand, it is important for you to know which character each reply is associated with, as trying to figure out who's talking from the text alone could get confusing in many scenes. So any format change should at least preserve the names.

Replies from: GWS, Yoav Ravid
comment by Stephen Bennett (GWS) · 2022-05-11T17:24:12.416Z · LW(p) · GW(p)

Would a play format work? e.g.

"Iarwain (languidly): I don't know that I am much interested in your family life, Lintamande"

Someone would have to translate the various profile pictures into emotions, but I would expect there to only be a handful of them per character.

Replies from: rocurley
comment by rocurley · 2022-05-14T17:50:04.914Z · LW(p) · GW(p)

Planecrash has 307 images at the time I made this comment, for context.

comment by Yoav Ravid · 2022-05-11T15:37:43.806Z · LW(p) · GW(p)

So perhaps the character names could be kept, maybe the pictures too (though I don't know if there's a good way to include so many pictures in an epub format, it's not exactly built for having small images beside the text, as far as I know), but not the authors names. Cause though it's nice knowing who wrote what, it's also confusing because everyone uses special usernames and I don't yet know which name is the author's name and which is the character's.

Replies from: benjamincosman, MondSemmel
comment by benjamincosman · 2022-05-11T17:22:56.936Z · LW(p) · GW(p)

The author's (user)name is always the final line of the inset box. If there are three lines in that box (e.g. "Carissa Sevar // to-let-you-in // lintamande"), then lintamande is the author, Carissa Sevar is the character, and "to-let-you-in" can be ignored (it's some sort of thematic tag for the character).

Authors other than the main two don't appear for an extremely long time so I'd worry about that after getting that far :)

comment by MondSemmel · 2022-05-11T16:11:29.407Z · LW(p) · GW(p)

"Iarwain" is Eliezer, "Lintamande" is his co-author, and IIRC that's basically it for most of the story.

comment by rocurley · 2022-05-13T09:41:52.249Z · LW(p) · GW(p)

If you don't want to bother using the glowfic downloader yourself, here's the epub so far. My intention is to update it every day, but no guarantees.

comment by RomanS · 2022-05-11T11:43:34.146Z · LW(p) · GW(p)

I'm also thinking about writing such a script. Beautiful Soup is the perfect lib for the job.

comment by Writer · 2022-05-12T10:25:35.355Z · LW(p) · GW(p)

One way to preserve the pictures and the general format while making the story more easily readable is simply converting the books into PDF. At least you could read them on a tablet, which is easier and less distracting than PCs. Tablets' screens are also much better for reading than PC monitors, even if they aren't as good as Kindles. It would be a good compromise, I think.

Replies from: Yoav Ravid
comment by Yoav Ravid · 2022-05-12T10:37:28.470Z · LW(p) · GW(p)

Tablets' screens are also much better for reading than PC monitors

Oh, really? What's the difference between them? I did notice that I like reading on my tablet better (I use it as my kindle) but I thought that it's because I'm using a black background.

Replies from: rhollerith_dot_com
comment by RHollerith (rhollerith_dot_com) · 2022-05-12T16:30:18.817Z · LW(p) · GW(p)

Small screens emit less light and consequently less of the blue light that tends to disrupt a proper sleep cycle than large screens do.

Yes, the smaller size of the tablet is cancelled out by the fact that people tend to hold it closer to the eye with the result that just as much of the light coming directly from the screen enters the eye, but in the case of the larger screen viewed further away, there is much more indirect light (e.g., light that has bounced off the walls of the room) entering the eye.

(As the amount of light entering the eye increases, the pupils constrict in compensation, but at the intensities used for reading at night, the compensation is only partial if my experience and careful observation of myself is any indication.)

Also, many on this site are nearsightedness (because nearsightedness is correlated with intelligence). Nearsightedness would be my guess as to the cause of his/her belief that "Tablets’ screens are also much better for reading than PC monitors".

comment by Raemon · 2022-05-11T23:04:56.741Z · LW(p) · GW(p)

I don't get much out of the glowfic format, but I'd be interested in posts that basically excerpt out the parts that are like "here's how Dath Ilan works and why it's good", or that impart particularly neat rationality/coordination lessons. Some people have already been doing that, but I wanted to flag my own appreciation.

Replies from: Pattern
comment by Pattern · 2022-05-12T02:22:43.980Z · LW(p) · GW(p)

here's how Dath Ilan works and why it's good

It does not seems like there will be straight answers to that:

It's the story of Keltham, from the world of dath ilan; a place of high scientific achievement but rather innocent in some ways.  For mysterious reasons they've screened off their own past, and very few now know what their prescientific history was like.

Replies from: thomas-kwa
comment by Thomas Kwa (thomas-kwa) · 2022-05-12T05:43:14.549Z · LW(p) · GW(p)

There's actually quite a bit about dath ilan exposed in planecrash-- they've screened off their history from their citizens, not from readers. Also many of the ways dath ilan is good doesn't depend on its history, and many of the rationality lessons in planecrash aren't directly about dath ilan.

There are a few principles [LW · GW] I'd be interested in people extracting, but two things I'd be particularly excited about (minor spoilers):

  • "Lawfulness" and its facets: Bayes, expected utility, the ability to coordinate and trade, etc.
  • How Keltham analyzes everything to try to understand it as an equilibrium between rational actors, whether this works in real life, and how to do it (partially covered in Inadequate Equilibria)
comment by shiney · 2022-05-11T12:16:09.582Z · LW(p) · GW(p)

Does anyone know if there's work to make a podcast version of this? I'd definitely be more willing to listen even if it is just at Nonlinear library quality rather than voice acted.

Replies from: planecrashpodcast, Henry Prowbell
comment by Henry Prowbell · 2022-05-11T15:57:51.745Z · LW(p) · GW(p)

I strongly agree

comment by Richard_Kennaway · 2022-05-11T17:42:25.635Z · LW(p) · GW(p)

How much of the story and world-building was developed in advance, and how much arose through the process of telling it? Do the co-authors currently plan things out together off-forum, or is what we read both the story and the process of creating it?

Replies from: MondSemmel, ADifferentAnonymous
comment by MondSemmel · 2022-05-11T21:24:04.643Z · LW(p) · GW(p)

As I understand it, the setting of Golarion and Cheliax is Lintamande's take on the world from the Pathfinder RPG system. She'd already written lots of glowfic stories in that setting by the time Planecrash came to be. Meanwhile, Eliezer had already written about dath ilan in some other places (like the original April Fool's post [LW · GW], and later on in some shorter glowfic threads). So there was definitely lots of pre-existing worldbuilding.

But whenever the story demands it, the world is further fleshed out. Eliezer even occasionally posts Twitter questions on things like "what's the best way to do X" as inspiration for dath ilan's Civilization.

comment by ADifferentAnonymous · 2022-05-12T12:49:48.133Z · LW(p) · GW(p)

Do the co-authors currently plan things out together off-forum, or is what we read both the story and the process of creating it?

I wonder this too. My impression is that it's some of both.

comment by Giovanni · 2022-05-18T22:31:23.331Z · LW(p) · GW(p)

Here is an alternative for generating the epubs, adapted from a version I wrote a few weeks ago. The format is similar to Glowfic, but with images/characters to the right. This way (imo) the text flows better, and the image hints at who is talking in the corner of the eye without having to read their name.

I would be up for setting up the planecrash posts to be added and auto-updated in the repo above for easy downloading if the authors want that.

A note for anyone else writing scripts: Glowfic has an API, you can check the linked code or their repo for specifics.

 

comment by Kenny · 2022-06-03T02:19:23.709Z · LW(p) · GW(p)

Thanks for the new story! It's great.

I suspect it's at least somewhat idiosyncratic, but I found myself, so late at night on the first day I started it that it was also very early morning, wishing that whomever had first pointed me at this had included a 'content warning' like what people (reasonably) often provide when linking to, e.g. TV Tropes. I found the story extremely engaging.

comment by RomanS · 2022-05-11T11:03:05.972Z · LW(p) · GW(p)

Thank you very much! 

As I understand, a part of the enjoyment from reading this kind of creative work is to observe the interactions between the co-authors. Thus, the forum format. 

Nevertheless, I predict that many readers (myself included) would prefer a plain-text book, devoid of the usernames, forum formatting etc. 

If there is a plain-text rendering, please kindly provide a link. If there is no such rendering yet, I think I may be able to automatically generate it.

Replies from: Dentin
comment by Dentin · 2022-05-11T11:31:57.587Z · LW(p) · GW(p)

I would prefer plain text, or at least dramatically more compact.  I find glowfic, including this one, to be borderline unreadable because of the format.

Replies from: Dirichlet-to-Neumann
comment by Dirichlet-to-Neumann · 2022-05-12T16:43:43.835Z · LW(p) · GW(p)

It's not just the format for me. Glowfic is really diluted and lack direction (as expected in a collaboration/improvisation). Consequently it tends to say in 10^6 words what could have been said in half as many.

I tend to think that when your novel is twice the length of the Lord of the Rings and counting it is probably a bit too long.

Replies from: Dweomite
comment by Dweomite · 2022-05-12T18:56:26.522Z · LW(p) · GW(p)

When something is a million words long, you should probably be thinking of it like a series rather than a single novel.

But it's not crazy for a series to be that long; for comparison, that's about the length of Harry Potter (the original 7-book series, not HPMOR).  A Song of Ice and Fire is up to around 1.8 million (and counting).

comment by niplav · 2023-12-21T17:02:40.428Z · LW(p) · GW(p)

I don't think this would fit into the 2022 review. Project Lawful has been quite influential, but I find it hard to imagine a way its impact could be included in a best-of.

Including this post in particular strikes me as misguided, as it contains none of the interesting ideas and lessons from Project Lawful, and thus doesn't make any intellectual progress.

One could try to do the distillation of finding particularly interesting or enlightening passages from the text, but that would be

  1. A huge amount of work[1], but maybe David Udell's sequence [? · GW] could be used for that.
  2. Quite difficult for the more subtle lessons, which are interwoven in the text.

I have nothing against Project Lawful in particular[2], but I think that including this post would be misguided, and including passages from Project Lawful would be quite difficult.

For that reason, I'm giving this a -1.


  1. Consider: after more than two years the Hanson compilation bounty [LW · GW] still hasn't been fulfilled, at $10k reward! ↩︎

  2. I've read parts of it (maybe 15%?), but haven't been hooked, and everytime I read a longer part I get the urge to go and read textbooks instead. ↩︎

comment by AprilSR · 2024-01-18T23:22:10.914Z · LW(p) · GW(p)

I feel like Project Lawful, as well as many of Lintamande's other glowfic since then, have given me a whole lot deeper an understanding of... a collection of virtues including honor, honesty, trustworthiness, etc, which I now mostly think of collectively as "Law".

I think this has been pretty valuable for me on an intellectual level—I think, if you show me some sort of deontological rule, I'm going to give a better account of why/whether it's a good idea to follow it than I would have before I read any glowfic.

It's difficult for me to separate how much of that is due to Project Lawful in particular, because ultimately I've just read a large body of work which all had some amount of training data showing a particular sort of thought pattern which I've since learned. But I think this particular fragment of the rationalist community has given me some valuable new ideas, and it'd be great to figure out a good way of acknowledging that.

comment by Aryeh Englander (alenglander) · 2022-05-11T13:28:58.414Z · LW(p) · GW(p)

Darn, there goes my ability to use Iarwain as a really unusual pseudonym. I've used it off and on for almost 20 years, ever since my brother made me a new email address right after having read the LOTR appendixes.

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-11T16:25:46.567Z · LW(p) · GW(p)

...sincere apologies.

Replies from: alenglander
comment by Aryeh Englander (alenglander) · 2022-05-11T19:24:26.657Z · LW(p) · GW(p)

Heh, no problem. At least I changed my LessWrong username from Iarwain to my real name a while back.

comment by Ben Pace (Benito) · 2024-01-17T21:17:18.114Z · LW(p) · GW(p)

Reading Project Lawful (so far, which is the majority of Book 1) has given me a strong mental pointer to the question of "how to model a civilization that you find yourself in" and "what questions to ask when trying to improve it and fix it", from a baseline of not really having a pointer to this at all (I have only lived in one civilization and I've not been dropped into a new one before). I would do many things differently to Keltham (I suspect I'd build prediction markets before trying to scale up building roads) but it's nonetheless extremely valuable to read someone's attempt at this.

The thing I dislike most about it is that every interaction is suffused with highly adversarial deceptive analysis. I find this pretty hard to do in real life and kind of distasteful and is not a skill I aspire to have. I understand Keltham finds himself in a highly adversarial environment, but I still don't like it.

I really wish it had chapters or similar units of chunking. I bounced off like 4 times before being able to read this book, having to learn the practice of "this is about enough reading for now / this is probably a good place to stop" which most books help me with themselves.

Overall +9 as an assessment of the quality of the contribution, though I agree that I'm not quite sure how this would fit in the review. Perhaps we could just include a select few pages of it for flavor then link to the website.

comment by NoriMori1992 · 2022-09-01T07:39:59.712Z · LW(p) · GW(p)

Love accidentally glancing at the content warnings at the top of the post and getting massive spoilers. Like these are seriously the most spoilery content warnings I've ever seen in my life. "Hey so here's some really specific info about some shocking stuff that happens later, which the beginning of the story doesn't hint at in the slightest." I'm not against the warnings existing, but could they at least be under some button or other barrier that says they're enormous spoilers?

Replies from: TheTrueSquidward
comment by Space L Clottey (TheTrueSquidward) · 2023-08-17T01:14:46.943Z · LW(p) · GW(p)

Agreed, the spoiler level was insane

comment by Yitz (yitz) · 2022-05-11T12:17:09.033Z · LW(p) · GW(p)

May I ask what the copyright status of this work is? This could be relevant if someone wants to publish a spinoff or something set in the same universe. (to be clear, I have no plans to do that, but it seems worth asking)

Replies from: Kenny, Kenny
comment by Kenny · 2022-06-03T02:11:59.549Z · LW(p) · GW(p)

I'm pretty sure Pathfinder (really Paizo), the creators of the world in which the story is set, has some kind of 'open license' for some of their content and I don't expect they're likely to object to this, given that it's basically 'fan-fiction'.

But I also don't think the copyright status is clear enough that anyone would want to "publish" derivative works – beyond something similar to this work itself.

comment by ClipMonger · 2022-12-02T09:16:04.767Z · LW(p) · GW(p)

I don't think that putting in the guide was a very good idea. It's the unfamiliarity that makes people click away, not any lack of straightforwardness. All that's required is a line that says "just read downward and it will make sense" or something like that and people will figure it out on their own nearly 100% of the time.

Generally, this stuff needs to be formatted so that people don't click away. It's lame to be so similar to news articles but that doesn't change the fact that it's instrumentally convergent to prevent people from clicking away. 

Replies from: Yoav Ravid
comment by Yoav Ravid · 2022-12-02T17:23:32.095Z · LW(p) · GW(p)

Contra: what pushed me away before isn't that it wasn't familiar but that I didn't get the format even after trying several times. That guide seems fantastic, though unfortunately I don't currently have the time to read the story.

comment by Charles Paul (charles-paul) · 2022-05-23T05:21:28.568Z · LW(p) · GW(p)

Just out of curiosuity, is “Lintamande” a member of the rationalist community in real life, is her (his? their?) identity known at all

Replies from: pktechgirl
comment by Elizabeth (pktechgirl) · 2022-05-28T08:14:01.659Z · LW(p) · GW(p)

It's one of those "many people know who it is but it definitely is not to be written down" deals.

comment by trevor (TrevorWiesinger) · 2024-01-15T16:08:40.904Z · LW(p) · GW(p)

In his dialogue Deconfusing Some Core X-risk Problems, Max H writes [LW · GW]:

Yeah, coordination failures rule everything around me. =/

I don't have good ideas here, but something that results in increasing the average Lawfulness among humans seems like a good start. Maybe step 0 of this is writing some kind of Law textbook or Sequences 2.0 or CFAR 2.0 curriculum, so people can pick up the concepts explicitly from more than just, like, reading glowfic and absorbing it by osmosis. (In planecrash terms, Coordination is a fragment of Law that follows from Validity, Utility, and Decision.)

This is the "glowfic" he was referring to.

I'm very sympathetic to Niplav and Max H's concerns that it's just way too long. However, I disagree with their thinking. The burden of the length has actually fallen much harder on me than most readers, and in spite of that, I still think that projectlawful is worth the read. 

The lessons from projectlawful are a package deal; if you want some, you have to read the rest, and you have to watch examples characters trying the lessons and sometimes failing and sometimes succeeding, and you have to care. The use of examples to approach each problem from multiple angles is a key element of what made Yud's original sequences [? · GW] great and what made the CFAR handbook [LW · GW] effective. 

Yudkowsky seems to have noticed that humans spend many hours a week reading fiction and zero hours a week becoming more rational, and that people weren't reading the sequences so much because it was "work". I currently think that both projectlawful and HPMOR were the correct moves to fix these problems.

Replies from: niplav, habryka4
comment by niplav · 2024-01-16T15:26:51.133Z · LW(p) · GW(p)

Thanks for the sort-of-response. My main point was that it's tricky to include Project Lawful in the review, since the past three times those were fairly short (and small!) books. Long things are fine, but I was critical about the value of including all of Project Lawful in the review.

comment by habryka (habryka4) · 2024-01-15T17:04:17.832Z · LW(p) · GW(p)

(that's a quote from Max H, not from me)

Replies from: TrevorWiesinger
comment by trevor (TrevorWiesinger) · 2024-01-15T17:26:33.215Z · LW(p) · GW(p)

whoops, corrected from habryka to Max H. That section was a bit unusual, this is not a normal mistake that I'm predisposed to making (this [LW(p) · GW(p)] is an example of a mistake that I'm very predisposed to making).

comment by Vanessa Kosoy (vanessa-kosoy) · 2022-05-15T09:47:08.955Z · LW(p) · GW(p)

Was Pathfinder only used for worldbuilding, or did you actually roll dice as part of generating the story?

Replies from: Eliezer_Yudkowsky, Raemon
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-15T17:25:27.732Z · LW(p) · GW(p)

We rolled dice on a few special occasions, but mostly it was worldbuilding.

comment by Raemon · 2022-05-15T19:18:25.920Z · LW(p) · GW(p)

Semi related question: If I want to read up on whatever the usual pathfinder setting is like before getting rationalfic’d, is there a particularly efficient and/or fun way to do that?

Replies from: SaidAchmiz, Kenny
comment by Said Achmiz (SaidAchmiz) · 2022-05-24T00:49:33.760Z · LW(p) · GW(p)

Doing so may seriously spoil your enjoyment of this story, as you see major things that Eliezer & co. fudged to make the story work.

Replies from: Raemon
comment by Raemon · 2022-05-24T01:06:50.456Z · LW(p) · GW(p)

Huh, can you say more? Is this different from seeing the ways HPMOR differed from Potter Canon?

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-24T08:22:59.945Z · LW(p) · GW(p)

I am not a connoisseur of Harry Potter, as I am of D&D/Pathfinder (22 years playing, 19 years DMing / designing / participating in the community), so I cannot be sure of the comparison’s validity, but—yes, I think it is different.

Project Lawful falls into that category of fiction (overwhelmingly, but not quite exclusively, represented by fan fiction) which can be roughly described as “person find himself in world which is governed by rules of tabletop roleplaying game, must now exploit those rules to survive/win”.

Such works vary, of course, but what is common to all of them is that in order to be… interesting, “valid”, “fair”… they have to, to the maximum extent possible, take as given the actual rules of the game in question, as they are experienced by actual characters when the game is actually played; and likewise, to the maximum extent possible, the fictional setting should be taken as it is exists in the actual game as played.

Why is this? Because the unique appeal of this genre, and the primary value of an instance of it, is similar to (but—IMO—much better than!) the value of a “whodunit” detective story: it presents a challenge for the main character(s), which also doubles as a challenge for the reader. “Being placed in this situation, and provided with this knowable, and known, set of tools, how do you win?” As the reader, you then get access to the dual enjoyment of watching the protagonist being clever and solving the challenge, and also getting to figure out how they should do it (in advance of reading about it). It’s an optimization game, in other words, where the constraints and the options are both much more well-defined than in most story genres, and also interesting and unusual.

This, however, can very easily be ruined, if the author alters the rules.

Now, some adjustment is unavoidable. It would be quite awkward, for example, to transfer the precise, grid-square-based, rules for attacks of opportunity into a textual fiction format, nor would this be desirable. In such a case, we can simply take for granted that the AoO rules are an abstraction that is meant to represent certain intuitively comprehensible realities of melee combat in a game-balance-preserving, tactically interesting, and battle-grid-compatible way, but are not intended to suggest that melee combat in Pathfinder “actually” has some sort of strange properties. The fic can simply present melee combat in a descriptive way (i.e., in the same way that a sword fight would be presented in one of George R. R. Martin’s works, for instance), because we have no need of that particular abstraction (the attack of opportunity system)—we are simply representing the thing being abstracted over, in a more appropriate way. This is all fine and not problematic at all; it’s just common sense; and so this is not the sort of rules “alteration” I am referring to.

I think my point will be best made with an example. I will not use an example from Project Lawful, for now (though of course I can provide such, on request). Instead, I offer what I consider to be one of the most egregious, and most thoroughly, comprehensively ruinous, instances of the problem I describe, which occurs in a story called Two-Year Emperor (or 2YE for short).

In 2YE, the protagonist finds himself conjured into a world that runs on D&D rules, in order to serve as emperor of a powerful nation for two years (hence the title). We get to see him perform various feats of cleverness as he exploits the rules and “munchkins” his way to victory, but his biggest, most impactful exploit comes when he figures out how to near-instantly “level up” arbitrary peasants and laborers and so on, transforming them into high-level fighters, wizards, etc. within minutes; he thus raises an army of incredibly powerful followers very quickly. Let me emphasize that this, if it were possible, would be an indescribably powerful ability in any D&D-like world; the first person to gain this ability would be catapulted into a position of insurmountable, crushing dominance over all rivals. It’s difficult to overstate the effectiveness of such a trick—if it were possible.

How does our protagonist do this? The logic goes like this: in D&D, you gain levels by gaining experience points; when your XP total reaches a certain value, you are level 2, a certain higher value—level 3, and so on. (So far, so acceptable.) And where does XP come from? Why, killing monsters! (Uh-oh! First red flag. But let’s roll with it.) The more powerful the monster killed, the more XP gained. So if a low-level person could somehow quickly kill some very high-level, very powerful monsters, he would inevitably gain a ton of XP. (Blatantly false as a claim of necessity, but… well, it gets much worse, so let’s hold objections for now.) So, for example, if you set up a giant mechanical monster kill box, where a caged monster sits bound within a deathtrap mechanism, and a person of no particular combat ability need only walk up and press a button to trigger the mechanism and slay the contained creature, that would count as defeating said creature, and would yield a giant pile of XP. (Absolutely, 100% not.) And if a person gained a huge amount of XP at once, they could gain a whole bunch of levels at once, and thus if a level 1 commoner killed enough ancient red dragons in one day, why then they could suddenly—poof!—become a 20th level wizard!

WRONG.

Why wrong? Let’s count the reasons:

  1. How, when, and how much XP to award is, in fact, entirely up to the DM. He could award XP only for slaying monsters; or he could also award XP for other things, like good roleplaying, or “quest completion”; or he could only award XP for those other things, and not for slaying monsters at all. It’s totally DM choice.

Ok, maybe the “DM” that runs this particular world has, in fact, chosen to award XP only for killing monsters (and sticks to this policy even in cases where it’s obviously insane—see below—something which no quarter-decent DM would ever do); and furthermore, this DM is in fact following the recommended (!) encounter XP award table in the Dungeon Master’s Guide. We also have to assume that the DM is choosing to ignore the suggestions that level acquisition carry with it a downtime/training requirement. Fine, let’s grant those assumptions.

  1. The aforesaid recommend encounter XP award table does not, in fact, have any entries for how much XP a level 1 commoner should get for killing an ancient red dragon. Indeed, the notes accompanying said table explicitly point out that if somehow a very weak character kills a very powerful monster, this basically has to mean that some sort of plot device, or some such, has happened to enable that, and so treating this as a successful exercise of the character’s capabilities (and thus deserving of experience points) is absurd.

  2. The Dungeon Master’s Guide also notes that if a character who supposedly “defeats” some powerful opponent has a lot of help from some person or circumstance, such that the effect of the aid far outweighs their own contribution, it likewise makes no sense to assign them XP for this “victory”. The “monster kill box” is as perfect an example of this as one could imagine.

  3. Finally, and most damningly for this “clever hack”, there is a specific rule that explicitly excludes the possibility of a character gaining multiple levels at once from an unusually large infusion of XP. This rule exists specifically to prevent precisely the kind of “clever exploit” that the protagonist of 2YE uses.

… but if we take a look at the author’s notes for Two-Year Emperor, we find a comment stating that the author is ignoring that “no multiple levels gained at once” rule, because he finds it to be “stupid”.

Well.

Remember this xkcd comic?

LIFEHACKS: You can just take all the luggage off the airport conveyer belt and leave with it. They don't check that it's yours at the door!

Imagine a story where in the preface, the author says: “In my fictional world, there aren’t any laws against insurance fraud; I think they’re stupid.” And then, in the course of the story, the protagonist, in an extremely impressive display of cleverness, carries out a brilliant get-rich-quick scheme: he takes out fire insurance on his decrepit house, then sets it on fire—and collects the insurance payout!

What would you think of a story like that?

That’s Two-Year Emperor, in a nutshell.

Now, there isn’t anything quite this egregious, in Project Lawful.

But (with apologies to Eliezer, because I am (mostly) enjoying the story, and don’t wish to convey the impression that I consider it to be irredeemably bad, by any means) I do think that there’s some stuff that is pretty close to being that bad.

And so, if you currently are sufficiently unfamiliar with the workings of the Pathfinder rules and the world of Golarion to not notice any such flaws, you may well prefer to remain thus ignorant.

Or not, of course. As I said—details available upon request.

Replies from: Eliezer_Yudkowsky, gjm, Raemon
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-25T23:09:34.946Z · LW(p) · GW(p)

Pathfinder rules-as-written seem merely unstable, in terms of how much they're munchkinable?  "Tears to Wine" in the hands of a 15th-level caster can provide a +10 competence bonus to Intelligence-based skill checks that lasts 3 hours?  Any wizard of any level with any magic item creation feat can transform 500gp of raw materials into 1000gp of magic item every 8 hours?

I don't think we've particularly been giving ourselves non-RAW useful-to-characters capabilities, with the exception of Prestidigitation chemistry as is a story conceit.  We've mostly been nerfing things because otherwise the world couldn't be in its depicted equilibrium.

I think I'd be interested in hearing about what you think is a violation of what we think is our principle.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-27T07:59:22.645Z · LW(p) · GW(p)

Hmm, I am not quite sure that I know what you’re asking. I don’t know what you think your principle is, so I couldn’t say what violates that. I will note what problems I perceive; whether they violate your principle is, of course, something you’re more qualified to determine than I am.


First, a slight digression on the subject of tears to wine. (You may skip this section if you are in a hurry, though I do think the point I make here is relevant to discussions of how “munchkinable” Pathfinder is.)

It is a well-known feature of 3e-like systems (D&D 3rd edition, D&D 3.5, Pathfinder) that they get more complex with time, as their creators release more and more “splatbooks” and other rules content. There is no real “availability scoping” in the rules, so when Paizo publishes a new book of rules content, such as the Arcane Anthology, there is no hard distinguishing factor between, say, a spell that appeared in the Core Rulebook and one that appears in the newly-published book. In this way, the amount of stuff in the game system increases monotonically with time, and likewise does the number of possible interactions between system components.

This makes “munchkining” a “mature” 3e-like system easier than doing so with a younger such system—there’s simply many more things to potentially exploit (and “power creep” is a thing as well; for reasons of market incentives, later-published content tends to be more exploitable than earlier-published content). This is a problem for users of the system, but it is not as much of a problem as it could be, because the solution to this, as to many other things, is the Game Master. It is commonly understood that a GM is well within his prerogative not to simply allow the use of all published theoretically-canonical game content, but to limit what is available, to one degree or another. (Indeed, you will hear this recommendation perhaps most clearly precisely from those communities of D&D/PF players who specialize in analysis of “theoretical optimization” a.k.a. munchkining.)

After all, if Paizo publishes the Arcane Anthology (where the spell tears to wine is found), there is not, actually, any law that says that this fact automatically means that any of the content in said “splatbook” is true in your specific campaign setting (if you are a GM). It’s your choice! Now, of course, you can take it as an axiom that all canonical published content is true of your setting (the Eberron campaign setting for D&D 3.5 is, famously, built on this premise). But you don’t have to do that.

And given that you don’t have to do it, making that choice is, well, a choice. By selecting “all published canonical content” as the scope for what is true in your Pathfinder-based setting, you inherently make it much easier for yourself, if you want to do “munchkinry”. With a 3e-like system as mature as 1st edition Pathfinder, when all canonical published content is “in scope”, coming up with an exploit is more often merely a matter of finding the right spell in the right splatbook (a task made much easier by websites such as the Archives of Nethys) than it is of any particularly clever hack.

Note that such a broad scope also substantially reduces the value of the work to the reader, along the dimension I describe in the grandparent comment. After all, if I am reading along and thinking “hm, how will the Conspiracy handle this one, let me think now”, then even if I am fairly well familiar with Pathfinder, there’s no way I can recall every feat, every spell, every magic item, every exotic piece of rules content in every splatbook, Adventure Path, and anything else that Paizo has ever published! With such a vast universe of possibilities, I have little choice but to resign myself to the fact that I have no idea what’s going to happen next. It could be anything at all. In Pathfinder 1st edition ca. 2022, “finding an obscure spell that does <thing>” is, in practice, little different from “making up, de novo and for your plot convenience, a spell that does <thing>”. You are technically remaining within the genre-standard set of constraints… but the reader is almost entirely incapable of predicting your moves, because those constraints are so loose.

However, none of this is the sort of thing I had in mind when I wrote the grandparent comment. This is merely a digression—which is now over.


The most obvious problem with Project Lawful (and one of the most severe, due to how often it appears) is that message does not work that way.

Let’s first review the message spell. It is a 0th level spell (cantrip), castable at will by a spellcaster of any of the listed classes, if known (for spontaneous casters) or prepared that day (for prepared casters). It affects up to 1 creature per caster level (e.g., 3 creatures when cast by a 3rd-level wizard), has a duration of 10 minutes per caster level (e.g., 30 minutes for a 3rd level wizard), and a range of of 100 feet plus 10 feet per caster level (e.g., 130 feet for a 3rd level wizard). (Note that when a wizard is referred to as “second-circle” in Project Lawful, this appears to mean that they are able to cast 2nd-level spells but no higher, meaning that they have either 3 or 4 wizard class levels, in Pathfinder terms.) It has verbal, somatic, and focus components (the F component is a piece of copper wire, though this is mostly irrelevant). The spell’s description reads thus:

You can whisper messages and receive whispered replies. Those nearby can hear these messages with a DC 25 Perception check. You point your finger at each creature you want to receive the message. When you whisper, the whispered message is audible to all targeted creatures within range. Magical silence, 1 foot of stone, 1 inch of common metal (or a thin sheet of lead), or 3 feet of wood or dirt blocks the spell. The message does not have to travel in a straight line. It can circumvent a barrier if there is an open path between you and the subject, and the path's entire length lies within the spell's range. The creatures that receive the message can whisper a reply that you hear. The spell transmits sound, not meaning; it doesn't transcend language barriers. To speak a message, you must mouth the words and whisper.

The procedure for using message to communicate thus looks roughly like this (we will assume a 3rd level wizard when specific numerical values are needed; modify as appropriate for casters of other classes and/or levels):

  1. As a standard action (the spell’s casting time), select up to 3 creatures, all of which are within 130 feet of you, all of which you can see, and to all of which you have line of effect[1], and point to each of them as you cast the spell.

  2. At any time within the next 30 minutes, you can whisper a message; the message will then be delivered to the target creatures.

  3. The targeted creatures whisper a reply, which is delivered to you.

As the act of casting message and the act of using an active message spell to actually send a message are distinct, let’s consider each separately.

Casting the spell

The following conditions (among others) obtain when you cast a message spell:

a. You must be able to see all targets on which you cast message. (General rules for spellcasting.) You cannot cast message on someone in a different room, or around a corner, or with your eyes closed, or if they’re invisible (and you have no means of seeing invisible things), etc.

b. You must have line of effect to all targets on which you cast message. (Ditto.) Any solid barrier whatsoever blocks line of effect. This is another reason why you can’t cast message on someone around a corner or in the next room, but the line of effect requirement also prevents you from casting message on someone on the other side of a transparent glass window, or a wall of force, etc.

c. You must provide all of the spell’s components—in this case, a verbal component, a somatic component, and a focus (a piece of copper wire). Relevant rules include:

A verbal component is a spoken incantation. To provide a verbal component, you must be able to speak in a strong voice.

A somatic component is a measured and precise movement of the hand. You must have at least one hand free to provide a somatic component.

A focus component is a prop of some sort. Unlike a material component, a focus is not consumed when the spell is cast and can be reused. As with material components, the cost for a focus is negligible unless a price is given. Assume that focus components of negligible cost are in your spell component pouch.

(That “negligible cost” provision does indeed apply to the piece of copper wire which is the focus for a message spell.)

To summarize: in order to cast message, you must speak an incantation in a strong voice; you must make a measured and precise movement of the hand; and you must provide, from your spell component pouch, a piece of copper wire as a prop.

An obvious question: are there any ways to avoid having to provide one or more of the spell’s components?

There are some such ways. Here is one which does not work: the Eschew Materials feat, which allows a spellcaster to ignore material components when such components cost less than 1 gp… but, unfortunately, does absolutely nothing about having to provide a focus, whatever its cost might be.

One might also use metamagic feats, such as Silent Spell and Still Spell; these allow a spellcaster to modify a spell so as to be castable without verbal or somatic components, respectively. (I am not aware of a metamagic feat that would let a spell be cast without a focus, if the spell normally has a focus.) However, such feats modify the spell’s effective level; so a wizard, e.g., would have to prepare message as a 1st level spell for it to be Still or Silent, or as a 2nd level spell for it to be Still and Silent. A message spell prepared thus would not be castable at will, as that is a property of 0th-level spells (cantrips) only.

d. You must point your finger at each creature on whom you wish to cast the message spell.

All of these things combine to create two important effects:

i. When message is cast, its targets—i.e., those creatures to whom the caster can, at any time in the spell’s duration, send whispered messages—are limited to those creatures who are in the same “room” (more broadly: between eligible targets and the caster there must be no solid barriers, not even translucent ones, nor can they be, in any way, hidden from the caster’s sight; nor can they be more than [100 + 10 per caster level] feet away).

ii. The act of casting message is—for low-level wizards such as the Project Lawful girls, anyway—very clearly noticeable by anyone in the vicinity.


Digression on spell manifestations (skippable if you are in a hurry):

Paizo’s official Pathfinder RPG Core Rulebook FAQ includes the following entry:

What exactly do I identify when I’m using Spellcraft to identify a spell? Is it the components, since spell-like abilities, for instance, don’t have any? If I can only identify components, would that mean that I can’t take an attack of opportunity against someone using a spell-like ability (or spell with no verbal, somatic, or material components) or ready an action to shoot an arrow to disrupt a spell-like ability? If there’s something else, how do I know what it is?

Although this isn’t directly stated in the Core Rulebook, many elements of the game system work assuming that all spells have their own manifestations, regardless of whether or not they also produce an obvious visual effect, like fireball. You can see some examples to give you ideas of how to describe a spell’s manifestation in various pieces of art from Pathfinder products, but ultimately, the choice is up to your group, or perhaps even to the aesthetics of an individual spellcaster, to decide the exact details. Whatever the case, these manifestations are obviously magic of some kind, even to the uninitiated; this prevents spellcasters that use spell-like abilities, psychic magic, and the like from running completely amok against non-spellcasters in a non-combat situation. Special abilities exist (and more are likely to appear in Ultimate Intrigue) that specifically facilitate a spellcaster using chicanery to misdirect people from those manifestations and allow them to go unnoticed, but they will always provide an onlooker some sort of chance to detect the ruse.

If this rule holds, then all spells (certainly including message) have obvious-to-onlookers magical manifestations (regardless of whether the spell has any components!). This would then be an additional reason why the casting of message would be unambiguously noticeable to anyone in the caster’s vicinity.

Official FAQ entries constitute canonical game content; thus, if you have decided that all published official content is “in scope” in your setting, then this includes material found in FAQ entries as well. However, this particular FAQ entry is a famously controversial one. When it was published, there was a sense, among many Pathfinder players and GMs, that the designers at Paizo were trying to “retcon” into existence a rule which has never existed in any rules text, even by implication. (It is true that illustrations in published Pathfinder materials almost invariably show some sort of glowing runes or flashing lights or some similar sort of visible effect whenever they depict a character in the act of spellcasting—but are we to take artistic choices to constitute statements of rules intent, when they are backed up by nothing, not even so much as a passing mention, in the text? —thus went the reasoning, among many.)

I thus do not hold it against you, Eliezer, that you ignore this particular rule, in Project Lawful. Nevertheless, diligence demands that it be mentioned.

Digression ends.


Using the spell

Supposing that you (a hypothetical 3rd-level wizard) have cast message, selecting up to 3 creatures as the targets, you now have a 30 minute period during which, at any time and as many times as you like, you can send whispered messages to all of those creatures. When you wish to do so, the following conditions (among others) obtain:

a. Only those of your targeted creatures who are within 130 feet of you will receive your whispered message; any that are further away will not receive anything.

b. Should there fail to be a path (all parts of which are within 130 feet of you, and no part of which is blocked by 1 foot of stone, or 1 inch of common metal, or a thin sheet of lead, or 3 feet of wood or dirt, or a magical silence effect) between you and any of your targeted creatures, that target will likewise receive nothing.

c. To transmit a message, you must mouth the words and whisper. (You cannot just form the words mentally; nor can you use ventriloquism-like techniques to whisper without mouthing the words; nor can you mouth the words silently—remember that “the spell transmits sound”! Note that this requirement still holds even if the spell is prepared with the Silent Spell and/or Still Spell metamagic feats, as those feats affect a spell’s components, not its effect.)

Note that this condition means that not just casting, but also using, a message spell, is something that cannot be done without risk of nearby people noticing. Of course it is possible for someone to not notice when a person right next to them is whispering and mouthing words… but it requires that the former person be somehow distracted, not paying attention, not looking directly at the message-user, etc. (In particular, it is not possible, for instance, for there to be three people in a room, all of whom are looking at each other and who are not intensely concentrating on something extremely distracting, and for two of them to be exchanging messages via message while the third is totally oblivious to this “side channel” and the conversation taking place on it.)

Likewise, it’s clearly not possible to use message to send messages while you are speaking aloud (your mouth and vocal cords can’t do two things at once, after all).

d. Those nearby can hear these messages with a DC 25 Perception check.

For reference, another hearing-based task that takes a DC 25 Perception check is “ Hear a bow being drawn”. This is difficult, but not impossible. (Note that this is the DC to hear the messages, not the DC to notice that someone is whispering something—the latter would require a check at a lower DC!)

Summary

Casting a message spell is very noticeable, and is limited by your locale’s layout in the targets available to you.

Using an active message spell to send messages is not as obvious, but still unquestionably noticeable to anyone in your immediate vicinity, and is less limited in applicable recipients by your locale’s layout, but still limited (and one must re-cast message if one wishes to add targets which were not selected by an already-active message spell).

But in Project Lawful, people routinely use message as, effectively, a fully telepathic side channel for unrestricted verbal communication which can be used without any risk whatsoever of someone who is directly adjacent to sender or recipient(s) noticing anything happening. (I have many examples which I can produce, but this comment is very long already. Details, as before, available upon request.)

It is clear that this usage is not supported by the rules text. But this is no mere quibble; the deviation is consequential. The ability to use message in this “basically just telepathy” way is a substantial boost to the capabilities of low-level spellcasters. (No sensible GM would ever permit message to be used in the way that it is used in Project Lawful, and with good reason—spellcasters, already quite powerful in Pathfinder, ought not be further empowered by misuse of the rules!)

Furthermore and specifically, if the members of the Conspiracy had to stick to the Pathfinder rules as written in their use of message, they could not perform many of the deceptions which they perpetrate upon Keltham. Substantial chunks of the story would either not work at all, or would have to be rewritten, sometimes from scratch.

Final note

The abuse of message is not the only problem of this sort in Project Lawful. It is, however, perhaps the most glaring one (at least, to me), and one of the most pervasive. This comment is, as I said, already very long, so I will forbear to list other examples—but more examples are, indeed, available upon request.

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-28T06:34:37.812Z · LW(p) · GW(p)

Yup!  I originally didn't understand how Message works very well.  Having misunderstood it, I played it consistently from there.

If you think this is Terrible then you're holding the story to a standard it's not particularly intended to meet.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-28T07:03:46.497Z · LW(p) · GW(p)

I admit, I’m curious to hear what standard the story is intended to meet / “what you think is your principle”.

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-29T02:36:06.208Z · LW(p) · GW(p)

It's meant to be reasonably hard fantasy, not necessarily conformant to Pathfinder canon because that doesn't describe a world in near-equilibrium relative to the smart people running around with +6 headbands of vast intelligence (that do exist in-universe), but once the characters see something it ought to go on being true.  Above all it's hard decision theory.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-29T04:52:03.059Z · LW(p) · GW(p)

Well… but in that case there’s still a problem: as I noted elsethread [LW(p) · GW(p)], the first use of message in the story (when Keltham first learns to cast it) does actually seem to be correct as per Pathfinder RAW. Uses later in the story are inconsistent with that one.

I certainly wouldn’t think to hold Project Lawful to a standard of conformance to Golarion setting canon; that would be somewhat silly, from a literary standpoint. But as far as mechanics go, if you’re trying to do “hard fantasy”, then, yeah, it does seem like there are flaws. The message thing is one; another is protection from [chaos/evil/good/law]. (Does the spell’s protection against mental control work only against mental attacks made by opponents of the targeted alignment, or all opponents? In Pathfinder it’s the former, and that is how it’s described in the currently most recent section of story, but earlier it is described in the latter way. I suspect this might be a case of one of the authors getting the 3.5 and PF versions of the spell mixed up, as its anti-mental-attack functionality was changed in PF to function in the alignment-limited way.) There’s more, but I haven’t been keeping meticulous track; those two inconsistencies are just the ones that jumped out at me.

More broadly, while I am not quite sure what you mean by “hard fantasy” (I can make the obvious inference from context and by analogy to “hard sf”, of course, but mapping that concept to fantasy, with magic and so on, seems non-trivial, though not impossible), I do think that aside from any questions of internal inconsistency, changes like “buffing” message in the way that you did are problematic. As I say upthread, this is a noticeable boost to the power of (at least) low-level spellcasters, relative to the PF RAW baseline. A world such as described in the story, and where anyone who can cast a cantrip effectively has at-will, robust, undetectable-by-bystanders telepathic communication with nearby targets of their choice, should look noticeably different from a world where all else is equal but message merely works the way that it does in Pathfinder. It does not seem to me that the world of the story is worked out with such a capability in mind from the start (which makes sense if this change was made accidentally midway through).

(And then, aside from all of this, there is one deviation from the Pathfinder rules that is so big and so bizarre that I genuinely can’t tell whether it’s deliberate or… some sort of very, very odd house rule / practice that I’ve never heard of… or what. But it’s not an inconsistency, at any rate…)

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-29T20:40:09.147Z · LW(p) · GW(p)

So the main thing I missed about Message was the chance of it being overheard.  Most of what you are reading as 'pseudo-telepathic' communication is usually a character having their mind read by Security running Detect Thoughts, and then those thoughts being relayed to others via Security using Message, rather than by characters Messaging each other.

I remember checking Protection pretty carefully at the time and I think at the time it blocked against all the mental control, not just mental control originating from the targeted alignment.  Possibly a rules change to PF2?  But if not, Keltham is still running Enchantment Foil at the time.

There's multiple big deviations from RAW; the main one I can think of that I homebrewed for this is 'oracles go with gods and a god can have at most one oracle'.  If that's not what you're referring to then I don't know what you're so coyly hinting about, and that kind of coy hinting is not something I find particularly pleasant.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-30T04:04:46.661Z · LW(p) · GW(p)

So the main thing I missed about Message was the chance of it being overheard. Most of what you are reading as ‘pseudo-telepathic’ communication is usually a character having their mind read by Security running Detect Thoughts, and then those thoughts being relayed to others via Security using Message, rather than by characters Messaging each other.

Yes, there is definitely some of that, but also cases where that can’t plausibly be happening. (Also, in cases where it is happening, there ought to be a noticeable communications lag, e.g.: think message -> Security reads via detect thoughts -> Security transmits to recipient via message. In some cases there are intervening walls, etc.—i.e. the characters are in different rooms—so there would then be the additional step “Security transmits to another Security via message”, which second Security then transmits to recipient.)

I remember checking Protection pretty carefully at the time and I think at the time it blocked against all the mental control, not just mental control originating from the targeted alignment. Possibly a rules change to PF2?

Indeed not. All functions of protection from [alignment] are alignment-limited in PF1, and always have been. (You can verify this, if you’re so inclined, by checking early printings [in PDF] of the Core Rulebook; if you don’t have access to such, feel free to PM me, and we can rectify that. But probably this is not important enough to go to any such lengths.) My best guess remains that you accidentally happened to look at the 3.5 version of the spell text.

But if not, Keltham is still running Enchantment Foil at the time.

True enough. (Of course, enchantment foil is only a +4 bonus to the save, not immunity… but there’s certainly no reason not to assume that that +4 bonus did happen to make the difference between success and failure, on that particular save.)

There’s multiple big deviations from RAW; the main one I can think of that I homebrewed for this is ‘oracles go with gods and a god can have at most one oracle’.

Ah, I don’t think I’d consider that a deviation from RAW, as such. It’s true that this is not at all Golarion canon, but I don’t think there’s actually any rules that forbid this from being the case in a setting, or even in Golarion as such. (I don’t think it’s a particularly consequential change from setting canon, either.)

If that’s not what you’re referring to then I don’t know what you’re so coyly hinting about, and that kind of coy hinting is not something I find particularly pleasant.

Apologies; it wasn’t my intent to “coyly hint”, only to avoid cluttering up the comment thread with what might not be of interest to you.

What I was referring to was the idea that gods can, e.g, bestow seven cleric levels on someone, or four oracle levels on someone else, etc., i.e. that a god (in Golarion) can decide to just give a mortal a bunch of character levels. This is definitely not how things work in Pathfinder, where one gains character levels when one gains experience points, and where it’s not clear what it even means for someone to spontaneously become, e.g., a 7th-level cleric without “leveling up” in the usual “acquire XP, gain character levels” way. (What is Ione’s base attack bonus, for example, or her base save bonuses? How many feats does she have, how many skill points? Do these questions have any meaning, even? It seems like they should, given the other references to [very close in-world analogues of] game mechanics, but who knows… Is there a connection between character level, class level, level-dependent benefits such as BAB/BSB/feats/skill points, or should we assume that there’s no such connection? If there is a connection and it’s the usual one, then is there consequently no connection between these numbers and what they normally represent, i.e. improvement of ability via training and practice? If a 16th-level cleric cast blasphemy and Ione were caught in the area of effect, would she be killed, or only paralyzed? etc., etc.)

(Of course, one could make the argument that as the gods are not given game-mechanical definitions in Pathfinder, we can therefore ascribe any powers we like to them without violating any rules, but then it’s not clear why we should expect any kind of predictable world at all.)

As I said, this isn’t an inconsistency, as such (at least, not definitely one, though it does seem hard to square with the other limitations that the gods are described as having with respect to their involved in the mortal world—but I wouldn’t lean too hard on that impression), it’s just… very, very strange, for multiple reasons. I would have to give more thought than I thus far have, to all the setting implications of this apparent divine capability. (Any potential conclusions I might draw would probably also be underdetermined by what we’ve seen in the story so far.) At the very least, I am fairly confident that Golarion as it is described in canon is built with the assumption that this is not a thing that the gods can do.

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-30T04:23:58.749Z · LW(p) · GW(p)

Well, that one is standard in lintaGolarian, not an innovation of ezerGolarion, and happens in an earlier continuity as well.  We've reinterpreted a lot of mechanics like that for reasons of "They are not actually living in an RPG and experience points are not actually a thing."  Spell durations go up continuously rather than in discrete jumps per level, similarly.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-30T16:56:13.114Z · LW(p) · GW(p)

Hmm… I am not quite sure how to take the “lintaGolarion” / “ezerGolarion” stuff (it doesn’t seem relevant? but possibly I am just not familiar enough with this terminology to get the implication)… but I think that perhaps I’ve not gotten my meaning across. Let me try again:

That the characters in Project Lawful are not actually living in an RPG and are not actually governed by literal game mechanics is clear enough. The same is almost to the same extent true of characters in an actual Pathfinder game, though! As mechanics in 3e-like systems, including Pathfinder, tend overwhelmingly to be associated, those mechanics do represent things that are ostensibly true from an in-world perspective.

With that in mind, here’s a concrete example. The blasphemy spell, which affects nonevil creatures, has an effect that is determined by the difference between the caster’s caster level and the hit dice of potentially affected creatures. If cast by, say, a 16th-level cleric, blasphemy will kill nonevil creatures of up to 6 hit dice (assuming they fail their Will save), but will only paralyze creatures of 7–11 hit dice (ditto).

We can accept that experience points are not actually a thing in-world, likewise “levels”, etc., but it remains the case that if an evil “eighth-circle” cleric walks up to Ione and casts blasphemy, and she fails her save, there does need to be an answer to the question: what actually happens to her? Does she die, or is she only paralyzed?

Of course you can evade this question by altering blasphemy to not be HD-dependent, or removing it entirely (but I think it’s been mentioned in the text already? but perhaps you could retcon that, if so); but then are you going to remove all HD-dependent or level-dependent effects that have discrete “breakpoints”? There are quite a few of those! Deciding to remove from Pathfinder all mechanics that force you to make determinations of what level a character is, or how many hit dice they have, etc., seems to me to commit you to making some rather substantial changes to the system (with non-trivial knock-on effects).

Let’s assume that you don’t make such sweeping changes, and in particular that you leave blasphemy unchanged. Well, we know that Ione is (at the start of the story, anyhow) a ~3rd-level wizard (again, we do not need to believe that “levels” are a real thing in-world, only that Ione’s relevant properties map, for the purposes of resolving interactions with spells such as blasphemy, to “3rd-level”, give or take a level). So if our hypothetical “eighth-circle” evil cleric walks up and casts blasphemy, and Ione fails her save, she instantly dies.

So far, so good. Now we read on, and see that Ione has been granted four oracle levels. Now if that same evil cleric walks up to Ione and casts blasphemy, and she fails her save, then… what? Does she still die (as would be the case if Ione gained the spellcasting ability[1] and class features of a 4th-level oracle, but did not gain any hit dice, nor increased in character level)? Or, is she now only paralyzed (as would be the case if Ione did gain character levels and hit dice)? Note, we’re still perfectly happy to concede that there’s no such thing as “character levels” and “hit dice” in-world, but “what actually happens to Ione in this in-world quite coherently describable scenario” does need to have some answer!

I can keep going, but I think my point should be clearer now (if not, by all means let me know and I’ll try to clarify further). Note that we can construct similar scenarios for the other questions I asked in my parenthetical, i.e. we can construct those questions in such a way that we’re asking about concretely describable, observable, in-world facts, rather than making reference to dissociated game mechanics. You can answer all of these, I’m sure; my point is only that however you answer them, it seems to me that you’ll end up with a setup which is, at least, very weird and not really anticipated by the Pathfinder system (and which is therefore likely to require unforeseen alterations, adjudication of unusual interactions, etc.).


  1. Do we actually see Ione or Pilar cast any oracle spells, by the way? It now occurs to me that I can’t recall such a case, so perhaps they only gained the class features and not the spellcasting? Or did I miss them using oracle spells? ↩︎

comment by gjm · 2022-05-26T19:34:42.316Z · LW(p) · GW(p)

It seems to me that Planecrasn is very much not aiming for the thing you say is central to the genre (and criticize it for not doing right):

it presents a challenge for the main character(s), which also doubles as a challenge for the reader. “Being placed in this situation, and provided with this knowable, and known, set of tools, how do you win?”

Yes, readers of Planecrash know (or can know, if they take the trouble to find out) the standard Pathfinder game rules. But the characters in Planecrash, most notably Keltham, do not. Keltham thinks he might be insider a fiction but it hasn't crossed his mind (nor does there seem to be any particular reason why it should, or for us to expect that it will) that he is inside a roleplaying game with known rules.

There absolutely is a class of fiction whose key feature is "person familiar with game X suddenly transported into an instance of game X", and such fiction can function the way you describe, but Planecrash is not one of those and I don't think it's fair to judge it by how well it succeeds in being one of those.

(I haven't read a lot of glowfic, but my impression is that (1) there is a substantial body of other glowfic that also takes place in a modified Pathfinder world, and (2) none of that involves protagonists familiar with the game. So, to whatever extent there are close precedents for what Planecrash does in that respect, they don't give reason to expect it to work that way.)

Replies from: SaidAchmiz, Slider
comment by Said Achmiz (SaidAchmiz) · 2022-05-26T20:17:40.267Z · LW(p) · GW(p)

I think you have substantially misread my point.

It is not necessary, for anything I said, that Keltham should “know” that “he is inside a roleplaying game”. (I don’t think it’s even true, in-story, that “he is in a roleplaying game”, so there isn’t anything for him to “know”, in that case.)

I did not say anything about ‘a class of fiction whose key feature is “person familiar with game X suddenly transported into an instance of game X”’. What I said was:

“person find himself in world which is governed by rules of tabletop roleplaying game, must now exploit those rules to survive/win”

I will also note, even aside from your mischaracterization of my point, that your protestations are substantially subverted by the fact that the characters in Project Lawful are constantly talking about people’s ability score values, their Sense Motive and Bluff and so on, etc. Yes, there are in-story explanations (for some of these things, less so for others), but that’s the point: this is a world with knowable rules, and people know them. Including Keltham, increasingly.

Replies from: gjm
comment by gjm · 2022-05-26T23:08:51.026Z · LW(p) · GW(p)

It certainly seems as if I have misunderstood what you wrote, whether or not I misread it. I also think that your interpretation of what I wrote isn't quite what I meant, but at least some of that is because I was sloppy in my wording, so let me first of all try to fix that.

When I said "inside a roleplaying game" I didn't mean literally inside an actual-game-being-played, I meant inside a system whose rules are those of a roleplaying game. (E.g., "inside the world of Pathfinder", as opposed to "inside an actual game of Pathfinder that someone is playing".) I think you took me to mean the latter rather than the former; I should have worded things so as to rule out that interpretation; sorry about that.

Another thing to get out of the way at the start: I don't understand your last paragraph; so far as I can see, nothing I wrote is invalidated at all by the fact that within the fiction there are things like numerical "intelligence" scores and "Will saves" and so forth. We know that those things correspond to game mechanics, but it doesn't look to me as if the way they're written requires (e.g.) that any of the characters see them as game-mechanic-like.

OK, on to what I guess is the main thing: my apparent misunderstanding of what you wrote. I've reread your original comment and, while your vigorous repudiation of what I took to be your meaning sure seems to indicate that I misunderstood it, I'm still failing to see how I misread; the only interpretations I can come up with that are consistent with what you're now saying seem to me to make less sense than my original interpretation, rather than more.

(Which, to be clear, I appreciate likely indicates mostly that I am still misunderstanding! But maybe what follows will clarify the nature of my misunderstanding in a way that lets you fix it.)

Consider the following three claims one could make· One: "In this sort of work, the point is for protagonist and readers to know, from the outset, what the rules are, and to try to figure out how to win given those rules." Two: "In this sort of work, the point is that there are rules that the protagonist and readers can figure out, and then they can try to figure out how to win given those rules." Three: "In this sort of work, the point is that the readers know the rules from the outset; the protagonist may have to figure them out, but then readers and protagonists are both trying to figure out how to win given the rules."

(Where "win" means achieving whatever goals seem appropriate; I don't mean literally winning a literal game, of course.)

I originally took you to be endorsing One. You object to this and (if I am now understanding right) say that the point is Two or Three or something in their vicinity: what is required is a world with knowable rules rather than a world with specific rules known to the readers and protagonist. I am not quite sure whether your position is more Two or more Three.

I thought you meant specifically One, because some of the things you said (so far as I can tell) make sense only as consequences of One and not of Two or Three. For instance:

they have to, to the maximum extent possible, take as given the actual rules of the game in question, as they are experienced by actual characters when the game is actually played; and likewise, to the maximum extent possible, the fictional setting should be taken as it is exists in the actual game as played.

If Two is the point, this doesn't make any sense. For Two, there have to be rules but it doesn't particularly matter whether they are the rules of any particular game. They could be entirely different rules, and Keltham and the readers would be figuring them out together.

If Three is the point, then that would be a reason why the rules should be specifically the rules of the particular game that the readers will recognize from reading the work. But Three introduces a big difference between what the protagonist is trying to do and what the readers are trying to do: the readers already know the rules and the protagonist is trying to figure them out. There's nothing wrong (I think) with that, but I don't see how to square it with another thing you wrote:

it presents a challenge for the main character(s), which also doubles as a challenge for the reader. “Being placed in this situation, and provided with this knowable, and known, set of tools, how do you win?”

This seemed to me (and still seems to me) to be saying that the protagonist(s) and reader(s) are facing the same challenge, namely the one you describe in quotation marks. But that's exactly what isn't (or needn't be) happening in case Three.

I can see one way to reconcile these things. The protagonist is typically spending longer in in-fiction time thinking about things, getting information, etc., than the reader is in real-world time, so it could make sense for the protagonist's challenge to be harder than the reader's. But I can't see the slightest reason why the specific way in which it should be harder is "readers know pretty much all the details of the specific system, protagonist has to work that out" rather than, e.g., "readers know that the specific system is something akin to / derived from X, protagonist doesn't have that extra information", which is exactly the situation we have in Planecrash: we know Keltham is in something like Pathfinder-land, and Keltham doesn't know that.

So I'm confused. If you meant One, then my original comment stands: Planecrash is plainly not trying to satisfy the constraints of One, and I don't see any reason why it should be. If you meant Two, then I don't see how you get from there to a requirement to match the rules according to which "the game is actually played" aside from obvious concessions for realism-versus-artificial-gameplay. If you meant Three, then I don't see why you implied that the readers' and protagonists' challenges are the same, and I don't see how you get from Three to a requirement to match the rules. And I don't see anything intermediate between One, Two and Three to which some corresponding intermediate objection doesn't apply.

So quite possibly you mean something that's outside the space spanned by One, Two, and Three. But I can't think what.

Oh, one other possibility, I guess. Maybe you mean pretty much Two, and your actual objection isn't "the rules are different" but "the rules are worse", the point being that it spoils the fun when e.g. the rules are too exploitable, because then there isn't really an interesting challenge any more. But then (a) I don't understand why what you actually wrote was that writers are obliged to use as nearly as possible the same rules as the game rather than (something like) rules no more stupidly exploitable than the ones in the game, and (b) in this particular instance it seems highly plausible that the changes made to the rules make them less stupidly-exploitable rather than more. (As per Eliezer's comment in this thread.)

It remains the case that the most reasonable interpretation I can find for your original comment -- meaning the one that makes your argument make most sense to me -- is the one I originally gave it, asserting something like claim One. Apparently this was wrong, and I'm sorry I got it wrong, but I'm failing to understand what I should have understood instead, and my best concrete attempts in the light of your reply -- see above -- seem clearly less reasonable rather than more.

I'm pretty sure my error, whatever exactly it is, isn't one of misreading. And I promise it isn't one of deliberate misrepresentation, which some of what you said in your reply seems to suggest! I'm sorry to be being so dim. What am I missing?

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-27T09:00:24.103Z · LW(p) · GW(p)

Consider the following three claims one could make· One: “In this sort of work, the point is for protagonist and readers to know, from the outset, what the rules are, and to try to figure out how to win given those rules.” Two: “In this sort of work, the point is that there are rules that the protagonist and readers can figure out, and then they can try to figure out how to win given those rules.” Three: “In this sort of work, the point is that the readers know the rules from the outset; the protagonist may have to figure them out, but then readers and protagonists are both trying to figure out how to win given the rules.”

(Where “win” means achieving whatever goals seem appropriate; I don’t mean literally winning a literal game, of course.)

I originally took you to be endorsing One. You object to this and (if I am now understanding right) say that the point is Two or Three or something in their vicinity: what is required is a world with knowable rules rather than a world with specific rules known to the readers and protagonist. I am not quite sure whether your position is more Two or more Three.

I certainly was not endorsing One.

Two, if that is entirely the approach that the author takes, puts the work into some other genre, so this is also not what I was talking about.

Three cannot quite be taken literally, as there are always facts about the setting that are not constrained by any game rules but are a “free choice” on the part of the author/GM, but that nonetheless constitute “rules” in the sense that they are stable facts about the world that the protagonist inhabits (and which he, at the very least, benefits from knowing). (Such things exist in almost all genres of fiction, of course.) Still, Three is the closest to my intended meaning.

I thought you meant specifically One, because some of the things you said (so far as I can tell) make sense only as consequences of One and not of Two or Three. For instance:

they have to, to the maximum extent possible, take as given the actual rules of the game in question, as they are experienced by actual characters when the game is actually played; and likewise, to the maximum extent possible, the fictional setting should be taken as it is exists in the actual game as played.

If Three is the point, then that would be a reason why the rules should be specifically the rules of the particular game that the readers will recognize from reading the work. But Three introduces a big difference between what the protagonist is trying to do and what the readers are trying to do: the readers already know the rules and the protagonist is trying to figure them out. There’s nothing wrong (I think) with that, but I don’t see how to square it with another thing you wrote:

it presents a challenge for the main character(s), which also doubles as a challenge for the reader. “Being placed in this situation, and provided with this knowable, and known, set of tools, how do you win?”

This seemed to me (and still seems to me) to be saying that the protagonist(s) and reader(s) are facing the same challenge, namely the one you describe in quotation marks. But that’s exactly what isn’t (or needn’t be) happening in case Three.

I do not agree with your analysis.

For one thing, what’s necessary is not that the readers already know the rules, it’s that (a) the readers know that the rules exist prior to the author writing (any particular part of) the story, so the author is working within a set of constraints that he isn’t just making up as he goes along according to his authorial whims (and also that the rules aren’t going to change mid-story, except by events in the story), and (b) the readers can, if they like, go look up the rules—if they want to figure out a solution to a problem the character is facing, or if they want to check whether a described solution works, etc.

(This is perhaps the distinguishing feature of this genre, in fact. In most works of speculative fiction, except perhaps for the hardest of “hard sf”, you really have no idea what the rules are, what they permit, what they forbid, etc. The author could spring something on you from nowhere at any time. He can declare that it works, and provide some contorted reasoning why it totally accords with previously-described things; and this is the case regardless of how impossible it would’ve been for you, as the reader, to either predict the new thing in advance, or to gainsay the author on the question of whether it all “really works”. That means that stories where the reader is challenged along with the protagonist are, in most subgenres of speculative fiction, basically farcical.)

Note that (a)—the readers knowing that rules exist prior to the story, and exist stably throughout the story (unless explicitly changed by story events)—does in fact mirror the protagonist’s situation. He, after all, is not in a story—he’s just in the real world, where the laws of reality do, indeed, exist prior to the protagonist endeavoring to figure them out, and exist stably, i.e. they don’t change mid-adventure (and if they do, it’s according to the stable and unchangeable real rules, which have no exceptions [LW · GW]).

Furthermore, there are two aspects to the appeal that I described.

First, there is the challenge of figuring out—in advance of reading onward to learn the author’s own answer—how the protagonist can, or will, solve whatever problem actually faces him. I am usually uninterested in attempting such challenges, because one can so very rarely trust authors enough to expect the challenge to be fair! For example, an author who takes any of the attitudes that you allude to, is obviously completely undeserving of the sort of reader trust that would make attempting to solve the “how can the protagonist win this” challenge anything but a waste of time. This is one aspect of the appeal of this particular genre of fiction—it makes such a level of reader trust possible.

But the second aspect of the appeal is the enjoyment of asking questions like “what could the protagonist have done instead?”, or “what should the protagonist do now?”, or “what could the protagonist’s opponents do, that would make the protagonist’s life more difficult?”, or “how might the story be different if the following conditions were tweaked?”—and so on in this vein.

Now, you can ask such questions of other sorts of stories, in other genres. But if you are dealing with speculative fiction of any sort, then such questions are usually pointless, because their answers amount to “whatever the author (or whoever is making the almost entirely arbitrary authorial decisions) wants”, “whatever the author (or etc.) decides will work”, etc. There’s no fact of the matter, so it’s all just fan-wank. (Many authors like to deny this, and claim that the rules they’ve allegedly laid out and allegedly followed constrain their story-worlds sufficiently that such questions are not devoid of meaning. I have little patience for such protestations; I find them to be absurd in nearly all cases.)

Not so in this genre. Here, we can ask the sorts of questions I listed above, and they can have answers, because the author is constrained. This is the key: the constraints upon the author, which are knowable by the reader.

Replies from: gjm
comment by gjm · 2022-05-27T12:40:19.829Z · LW(p) · GW(p)

It seems to me that you have picked a particular set of goals and constraints, called them "this genre", and declared your dissatisfaction with the fact that Planecrash is not seeking those goals under those constraints.

I agree that Planecrash does not appear to be seeking the goals you specify under the constraints you specify.

I don't see any particular reason to think either that its authors intend it to be doing those things, or that in some moral sense they ought to be doing those things. As for whether the work would be better if they were doing those things, I don't know, but it does seem like there's a substantial community of readers who are enjoying it without them.

(My feeling is that they are to some extent aiming  to be able to set challenges to the protagonist with corresponding challenges to the reader -- indeed there is concrete evidence that they are doing that at some particular points in the story -- but for that there is no need for the readers to know more about the rules than the protagonist does, and indeed I think it's better if their knowledge is pretty much the same as his.)

I do agree that there is a lot to be said for having rules that don't get made up ad hoc as the story progresses (even when the story isn't mostly about setting well-posed puzzles for the readers), and of course "non-speculative" fiction has that feature automatically, the rules being those of the world we're actually in. To whatever extent the point of the story isn't mostly to give readers well-posed puzzles, rules made up ad hoc are just as good provided they end up being as consistent and "fair" a set of rules as they would have been if fully worked out beforehand rather than e.g. being tweaked to get particular author-desired outcomes in particular situations. (One could e.g. imagine an author who, in the interests of fairness, decides that when some situation comes up that involves questions they haven't already decided the answer to, they will ask an impartial friend what the most natural way is for the rules to work, without telling them any plot information that might bias their choice. Maybe that would work well, maybe not, but it wouldn't have the sort of unfairness I think you are concerned about.)

I agree that the authors of Planecrash are clearly not working with exact Pathfinder rules. I don't see any strong evidence either way for whether they worked out all the rules in advance, nor for whether if they didn't they're choosing rules in a "fair" way when they run into something they hadn't pre-decided.

The specific major divergence you mention in another comment -- how very easy telepathic communication seems to be for the Cheliax folks -- is a thing that has bothered me too. I'm not sure whether that's actually an argument on my side of this present discussion or yours, though, because I don't know the Pathfinder rules about this and so my concern was "it's hard to believe that all these people with good-but-not-superhuman brainpower are able to do all this telepathic communicating while other things are going on too, without making it obvious" rather than anything along the lines of "oi, not fair, that isn't in the rules". It seems to me there's a roughly-equally-plausible possible world where the Pathfinder rules actually make Message overpowered like Planecrash-Message is, and in that possible world it would be better for the Planecrash authors to be more willing to diverge from the original rules, and I haven't seen anything suggesting that Pathfinder rules are so carefully balanced that I should expect a policy of not diverging from those rules to result in a more plausible world than a policy of diverging from them where doing so seems to make for a better story.

Planecrash needs the Project Lawful management to be able to deceive Keltham rather effectively. If their telepathic communication were much worse then the jig would be up within days. It is not clear to me that the story would have been improved by forbidding the authors to make Message more powerful; perhaps that would just have made the overall shape of the plot impossible, or perhaps they'd have needed more munchkinning on other things that are in Pathfinder rules and it would have ended up more implausible rather than less, to anyone who isn't a Pathfinder expert. (Maybe it would have forced them onto different paths that would have ended up making a better story. But I don't see any reason to think it would have.)

Anyway. Clearly Planecrash is not playing by the specific rules you would prefer it to be playing by. If you say that this greatly reduces your enjoyment of it, then of course I believe you. But if you say it's an objective failure to meet The Standards of The Genre, then I say: I just don't think it's trying to do the things you say it should be trying to do, and I don't see why it should be, and I'll match your declaration that it doesn't work for you with my declaration that so far as I can tell it works at least as well for me as it would if the authors were constrained to match actual Pathfinder rules exactly and I strongly suspect it would be worse for me if they were so constrained.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-27T20:48:04.021Z · LW(p) · GW(p)

Once again, I disagree with your characterization and your analysis.

It seems to me that you have picked a particular set of goals and constraints, called them “this genre”, and declared your dissatisfaction with the fact that Planecrash is not seeking those goals under those constraints.

You are speaking as if the classification of a story into a genre is something arbitrary—a personal choice I have made, according to my preferences and whims, and no more. This is inaccurate. Story genres are patterns of what sorts of structure and features of a work are enjoyable to readers, what sorts of reader responses they evoke, how effectively they do so, and other things of that sort. How we slice up the space of possibilities along these dimensions is certainly something we can debate; but it is not a matter of merely “picking and declaring”. When I say that this story falls into a genre, I am making reference to objective (or, at least, intersubjectively consistent) facts, not simply expressing my preferences.

I don’t see any particular reason to think either that its authors intend it to be doing those things, or that in some moral sense they ought to be doing those things.

What the authors intend, I cannot but speculate on (until such time as Eliezer & co. make their intentions known). That having been said, while I would not wish to push the “death of the author” line too far, still the author of a work does not have complete freedom in determining what sort of work it is. Likewise, genre conventions should certainly not be straitjackets (or else genres would never change, nor new ones arise); but deviations from such conventions ought to be justified by making a work better than otherwise (and, as in most such cases, it is almost invariably better for a work’s quality if the author understands the conventions he is violating, and commits the violations deliberately and with intent; note, by the way, that Eliezer absolutely does do this in other aspects of Project Lawful—see below).

I do not think morality, as ordinarily understood, enters into this, so at any time in this discussion when I say “ought”, I refer to aesthetic or prudential considerations, not ethical ones.

As for whether the work would be better if they were doing those things, I don’t know, but it does seem like there’s a substantial community of readers who are enjoying it without them.

Yes. You will recall that this conversation began when I advised someone that they might wish to avoid learning more about the Pathfinder rules, lest their enjoyment of the work be diminished.

Also, as I said earlier, I am enjoying the story myself. I have never said anything like “if Project Lawful hewed to the rules I set out, it would be good; failing to do so, it is bad”. That would be a false and foolish claim. But there are greater and lesser degrees of enjoyment; there are multiple dimensions on which a work may be appreciated, and a work may do worse on some dimensions than on others; a work may offer a more complete, more comprehensively satisfying reader experience, or, conversely, one that is flawed, marred by annoyances, unsatisfying in some ways.

This all seems like very basic media criticism to me. Frankly, “some people like it despite these so-called problems you describe!” is almost always a facile reply in such cases.

(My feeling is that they are to some extent aiming to be able to set challenges to the protagonist with corresponding challenges to the reader—indeed there is concrete evidence that they are doing that at some particular points in the story—but for that there is no need for the readers to know more about the rules than the protagonist does, and indeed I think it’s better if their knowledge is pretty much the same as his.)

Note that even if we take at face value what you say in this paragraph, it absolutely does not follow either that the authors shouldn’t avoid violations of the Pathfinder rules, or that readers do not benefit from being able to look up the rules. This is because if all that the reader knows about the rules is what the story tells us through the protagonist’s eyes, then the result will be that the reader knows less about the rules than the protagonist does! (This is the reason for the “character knowledge” concept in TTRPGs, and “Knowledge skills” in 3e-like systems, in particular. I hope that the logic of this is obvious enough, but I’ll explain it if needed.)

What’s more, even on this premise, the use of message is a flaw in the story. How do we know this? Because the first explicit use of the spell in the text does in fact acknowledge those facts about the casting and use of message that I described elsethread! It seems that the authors knew about this, but then, more or less, forgot (probably more like: didn’t really think about the implications, in their enthusiasm to use a very convenient plot device).

To whatever extent the point of the story isn’t mostly to give readers well-posed puzzles, rules made up ad hoc are just as good provided they end up being as consistent and “fair” a set of rules as they would have been if fully worked out beforehand rather than e.g. being tweaked to get particular author-desired outcomes in particular situations.

I completely disagree.

Rules made up ad hoc are sufficient to make an enjoyable story (as they would have to be, or most of the vast swath of works we call “speculative fiction” would be worthless). It is absolutely not the case that they are just as good as rules fully worked out beforehand.

I agree that the authors of Planecrash are clearly not working with exact Pathfinder rules. I don’t see any strong evidence either way for whether they worked out all the rules in advance, nor for whether if they didn’t they’re choosing rules in a “fair” way when they run into something they hadn’t pre-decided.

See above re: initial vs. later appearances of the message spell in the story.

The specific major divergence you mention in another comment—how very easy telepathic communication seems to be for the Cheliax folks—is a thing that has bothered me too. I’m not sure whether that’s actually an argument on my side of this present discussion or yours

Mine, and here’s why:

It seems to me there’s a roughly-equally-plausible possible world where the Pathfinder rules actually make Message overpowered like Planecrash-Message is

This is, in fact, one of those cases where what seems “plausible” to you is an artifact of your lack of knowledge, and the plausibility is defeated by greater knowledge of the system. (Recall a similar dynamic in the argument about philosophical zombies.)

It is not nearly as plausible (not totally unimaginable, mind you! but definitely much less plausible) that message should be that overpowered, if one knows more about 3e-like systems in general and Pathfinder in particular. Importantly, the specific information needed to reach this conclusion is available in the text of Project Lawful itself; the “greater knowledge” I refer to is general knowledge about the structure and patterns of such systems / this system. Namely:

Observe that there exists a higher level spell called telepathic bond (5th-level for wizards). This spell is mentioned in the story, and it is made clear that it’s more powerful, and less easily available, than the ubiquitous message cantrip. But why should such a spell exist, and why should anyone ever have use for it, if message is as powerful as the story makes it out to be? This is a strange design flaw for a system where the relative power and usefulness of similar categories of spells is strongly correlated with their relative levels! (What’s more, as both spells appear in the Core Rulebook—indeed, they were both inherited from the earliest iteration of this sort of system, D&D 3.0, where they both also appeared in the core rules—so this apparent anomaly is not even explainable by “power creep”, by splatbook option growth, etc.)

This is, again, not an insurmountable objection. It is not unknown for such “power ladder anomalies” to exist in 3e-like systems. But they are the exception, not the rule. Therefore message being as powerful as how it is depicted in Project Lawful is not totally implausible, but it is definitely substantially less plausible than message being the way that it is in the Pathfinder rules as written.

I haven’t seen anything suggesting that Pathfinder rules are so carefully balanced that I should expect a policy of not diverging from those rules to result in a more plausible world than a policy of diverging from them where doing so seems to make for a better story.

The Pathfinder rules are carefully balanced in some ways, less carefully balanced in other ways. Likewise, some deviations from the rules are clearly good and even necessary, while others are just as clearly false moves in the game of story design.

Let me offer an example of the former sort of deviation, for contrast; I think this may help you to understand how I am thinking about this. Eliezer wrote [LW(p) · GW(p)]:

Any wizard of any level with any magic item creation feat can transform 500gp of raw materials into 1000gp of magic item every 8 hours?

And this is entirely correct. The Pathfinder magic item creation rules famously differ from the magic item creation rules in D&D 3.5 (Pathfinder’s predecessor system) by removing the requirement that the creator expend his own experience points to make the item, and not replacing it with any other requirement for the expenditure of any other limited resource; thus the only limited resources that a character must expend to create magic items is money and time—but the exchange rate is such that… well, it’s as Eliezer said.

This is a problem with the Pathfinder magic item creation system, and a problem with Pathfinder generally. It is solved in various ways by various GMs, the details of which we need not concern ourselves with, except to note that such solutions tend to be unsatisfactory from a world-building perspective even if they’re adequate from a gameplay perspective.

In Project Lawful, the solution to this problem is, of course, spellsilver. We (the readers) have not yet learned very much about this material, but whatever it is, it’s evidently a limited natural resource, which is expended in the creation of magic items. (Unsurprisingly, this makes spellsilver extremely valuable, etc.; the implications are obvious, and depicted in the story in ways that make sense.)

When “spellsilver” first appeared in the story, my immediate reaction was “Ah! Eliezer has chosen to solve the ‘unlimited item creation’ problem by introducing a limited natural resource requirement! Very good, a classic solution and a sensible one; let’s see where this goes.” A reader unfamiliar with the Pathfinder system would, presumably, instead merely take “spellsilver” as a comprehensible feature of the fictional world, not noting anything amiss; a Golarion where there is such a thing as spellsilver, a rare material that is needed to make magic items, is perfectly coherent and unproblematic from the standpoint of constructing, and understanding, a fantasy setting.

In short, the introduction of “spellsilver” is a deviation from the Pathfinder rules as written that is (a) obviously very deliberate, (b) done with full knowledge of the rules as they are, and clear intent to achieve a specific effect by the change, and (c) successful at both the particular goal of changing the setting in the desired way and, more broadly, at improving the setting and the story that may be written in it.

The alteration to the functionality of the message spell appears to have none of those properties.

Planecrash needs the Project Lawful management to be able to deceive Keltham rather effectively. If their telepathic communication were much worse then the jig would be up within days. It is not clear to me that the story would have been improved by forbidding the authors to make Message more powerful; perhaps that would just have made the overall shape of the plot impossible, or perhaps they’d have needed more munchkinning on other things that are in Pathfinder rules and it would have ended up more implausible rather than less, to anyone who isn’t a Pathfinder expert. (Maybe it would have forced them onto different paths that would have ended up making a better story. But I don’t see any reason to think it would have.

Frankly, this reads to me like a post-hoc excuse for shoddy work—and, what’s more, an entirely unnecessary one. I have rather greater confidence than you, it seems, in Eliezer’s ability to write a story that doesn’t involve this sort of carelessness with constraints. And I have little doubt that it would be a better story overall.

As for your last paragraph…

I have addressed the matter of fiction genres earlier in this comment. The “well I like it fine the way it is!” and “it would be worse if they fixed these flaws!” stuff seems like rationalization, at best. With respect, I detect a note of defensiveness in your responses. There is no need for that. A flaw is a flaw, and we can admit that without giving up our appreciation for those aspects of the story which are good (and there are, indeed, plenty of things about Project Lawful which are not only good, but good in very uncommon ways).

Who was it, after all, who once wrote: “if you can’t criticize, you can’t optimize”?

Replies from: gjm
comment by gjm · 2022-05-27T22:20:18.671Z · LW(p) · GW(p)

Classification into genres is not arbitrary. But I don't think inferences of the form "this story has features X, Y, and Z, which put it in genre G, and stories in genre G generally try to do P, Q, and R; therefore this story should be doing P, Q, and R, and not doing so constitutes a defect" are valid unless either you can get from XYZ to PQR without going through G, or XYZ are tied to G so strongly that no reasonable person familiar with G could deny that a story with XYZ is trying to do what stories in genre G generally try to do.

If Planecrash ends up being enjoyable to read, or instructive, or funny, or prophetic, or suitable for turning into a wildly successful 30-hour-long opera cycle, those merits are not in any way nullified by its not doing particular things that stories sharing some of its features often do.

I think the discussion is devolving into "Said states his position. Gareth states his position. Said restates his position. Gareth restates his position." which is seldom productive, so I shall leave it here unless there are particular things you are anxious to have a reply to (in which case, let me know and I'll probably oblige).

Regarding your final few paragraphs: it is possible that I am being defensive, but I am pretty sure I am not being defensive on behalf of Planecrash, which I certainly don't regard as flawless or unworthy of criticism. I simply disagree with some particular claims you are making about it, and if I am defending anything it is my position on what makes a given work of fiction better or worse.

comment by Slider · 2022-05-27T10:02:02.107Z · LW(p) · GW(p)

"convention" vs "rules" but Keltham has had some stuff in that direction crossing his mind.

comment by Raemon · 2022-05-24T08:38:40.939Z · LW(p) · GW(p)

Gotcha. (In my case, I was pretty disinterested in this aspect of Planecrash and mostly find myself skipping past the various D&D minutia. The thing I was interested in re: pathfinder was more about the story than the rules, i.e. how various gods and alignments and nations are normally portrayed)

It's not obvious to me how much Planecrash was (implicitly?) promising to deliver on the sort of thing you describe here, but, makes sense to be sad if you were hoping for that, and/or to caution people about it.

Replies from: SaidAchmiz, Raemon
comment by Said Achmiz (SaidAchmiz) · 2022-05-24T09:05:25.087Z · LW(p) · GW(p)

I will note that the Project Lawful examples I have in mind can’t be fairly described as “minutiae”; they get at fairly fundamental aspects of the structure of the rules and the world.

(That said, it’s of course quite reasonable not to care at all about that sort of thing.)

Replies from: Richard_Kennaway
comment by Richard_Kennaway · 2022-05-26T21:12:35.527Z · LW(p) · GW(p)

I'm somewhat familiar with D&D, having played it a bit and read the rulebooks about 40 years ago and not since. So I recognised at once that Planecrash was set in a D&D-like world. The system of alignments, character classes, ability scores, and levels, and so on were familiar. I had never heard of Pathfinder but now know that it's the specific D&D-like rules and world in which Planecrash is set. Beyond that I don't care what the exact rules are (of Pathfinder or D&D).

From that point of view, are any of the defects you have in mind still recognisable as defects?

The story of the munchkin 2-year emperor is clearly a failure even for someone with no knowledge of the specific D&D rules that it violates. The fact that the author flouted several rules that were in the way of his plot just makes it worse, it turns it into tennis with the net down, but the plot as it stands was already pretty bad.

Replies from: SaidAchmiz
comment by Said Achmiz (SaidAchmiz) · 2022-05-27T08:02:33.018Z · LW(p) · GW(p)

From that point of view, are any of the defects you have in mind still recognisable as defects?

Please see my reply to Eliezer [LW(p) · GW(p)] for one example of the sort of thing I have in mind.

comment by Raemon · 2022-05-24T08:40:35.958Z · LW(p) · GW(p)

(that said, your comment prompts me to wonder about the domain of videogames, where there typically isn't DM judgment restricting XP gain. You've also written about World of WarCraft and I find myself curious if the 2 year emperor trick works there)

Replies from: FeepingCreature, SaidAchmiz
comment by FeepingCreature · 2022-05-26T19:25:56.674Z · LW(p) · GW(p)

Most multiplayer games have some way to limit XP gain from encounters outside your difficulty, to avoid exactly this sort of cheesing. The worry is that it allows players to get through the content quicker, with (possibly paid) help from others, which presumably makes it less likely they'll stick around.

(Though of course an experienced player can still level vastly faster, since most players don't take combat anywhere near optimally to maximize xp gain.)

That said, Morrowind famously contains an actual intelligence explosion. So you tend to see this sort of stuff more often in singleplayer, I think. (Potion quality triggers off intelligence. Potions can raise intelligence.)

And of course the entire genre of speedrunning - see also, (TAS) Wildbow's Worm in 3:47:14.28(WR).

comment by Said Achmiz (SaidAchmiz) · 2022-05-24T09:03:49.981Z · LW(p) · GW(p)

It absolutely does not, precisely for reasons #2 and #3 I listed.

Game designers—especially designers of extremely popular games with large budgets and armies of playtesters—do not, as a rule, tend to be idiots. And they would have to be very stupid indeed to allow a very serious and very obvious exploit that has, furthermore, been known for decades.

This is why I find things like the aforesaid Two-Year Emperor “exploit” to be insulting to my intelligence as a reader. There’s few quicker ways, than that, to ruin any possible enjoyment of a story.

comment by Kenny · 2022-06-03T02:13:55.626Z · LW(p) · GW(p)

Like with lots of things like this, there's a dedicated wiki:

comment by NoriMori1992 · 2022-10-07T06:12:48.387Z · LW(p) · GW(p)

I found the sandbox thread but hurting people is wrong, and found the part about Quiet Cities, and I nearly cried because I can't describe how badly I want something like that and would move to one immediately if it existed.

comment by Rene de Visser (rene-de-visser) · 2022-05-11T13:44:07.356Z · LW(p) · GW(p)

I'd also like an EPUB version that is stripped as possible. I guess it might be necessary to prepend the characters name to know who is saying what, but I find the rest very distracting. I find it makes it hard to read.

comment by Aorou (Adnll) · 2022-09-23T12:59:23.445Z · LW(p) · GW(p)

I'm looking for the discord link! It was linked at some point, but I was catching up then, so I didn't want to see spoilers and didn't click or save the link.
But now I'd like to find it, and so far all my attempts have failed. 

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-09-27T02:31:57.265Z · LW(p) · GW(p)

Here's one https://discord.gg/45fkqBZuTB

Replies from: Adnll
comment by Aorou (Adnll) · 2022-09-28T16:30:21.942Z · LW(p) · GW(p)

Thank you :))

comment by Celenduin (michael-grosse) · 2022-08-20T19:38:14.399Z · LW(p) · GW(p)

Is there a dedicated Wiki (or "subject-encyclopedia") for Project Lawful? I feel like collecting dath ilan concepts (like multi-agent-optimal boundary) might be valuable. This could both include an in-universe summary and context of them, and out of universe explanation and references to introductory texts or research papers if needed.

comment by artifex0 · 2022-05-13T02:20:40.408Z · LW(p) · GW(p)

There are a lot of interesting ideas in this RP thread.  Unfortunately, I've always found it a bit hard to enjoy roleplaying threads that I'm not participating in myself.  Approached as works of fiction rather than games, RP threads tend to have some very serious structural problems that can make them difficult to read.

Because players aren't sure where a story is going and can't edit previous sections, the stories tend to be plagued by pacing problems- scenes that could be a paragraph are dragged out over pages, important plot beats are glossed over, and so on. It's also very rare that players are able to pull off the kind of coordination necessary for satisfying narrative buildup and payoff, and the focus on player character interaction tends to leave a lot of necessary story scaffolding like scene setting and NPC interaction badly lacking.

If your goal in writing this was in part to promote or socially explore these utopian ideas rather than just to enjoy a forum game, it may be worth considering ways to mitigate these issues- to modify the Glowfic formula to better accommodate an audience.

The roleplaying threads over at RPG.net may provide some inspiration.  A skilled DM running the game can help mitigate pacing issues and ensure that interactions have emotional stakes.  Of course, forum games run with TTRPG rules can also get badly bogged down in mechanics.  Maybe some sort of minimalist diceless system would be worth exploring?

It could also help to treat the RP thread more like an actual author collaboration- planning out plot beats and character development in an OOC thread, being willing to delete and edits large sections that don't work in hindsight, and so on.  Maybe going through a short fantasy writing course like the one from Brandon Sanderson with other RP participants so that everyone is on the same page when it comes to plot structure.

Of course, that would all be a much larger commitment, and probably less fun for the players- but you do have a large potential audience who are willing to trade a ton of attention for good long-form fiction, so figuring out ways of modifying this hobby to better make that trade might be valuable.

Replies from: Eliezer_Yudkowsky
comment by Eliezer Yudkowsky (Eliezer_Yudkowsky) · 2022-05-13T07:20:14.173Z · LW(p) · GW(p)

We are both experienced authors not in need of this advice at this level.

Replies from: kevin-brannan
comment by Kevin Brannan (kevin-brannan) · 2022-05-22T03:03:30.210Z · LW(p) · GW(p)

What level is that, Eliezer?

comment by MondSemmel · 2022-05-11T08:10:43.835Z · LW(p) · GW(p)